[User 421137]

I have the same problem (now).
Basically I have some problems with the scrapers (wrong language, finds nothing, wrong movie) but also with wrong data in movie DBs.
Sometimes the TMDB scraper is better, but typically the Universal Movie Scraper is better (and fits more my needs). Unfortunately you cannot decide which scraper can be used on movie level (only on library level).

The concept of "Combination NFO file" is great, but it's not working well.

I get "Found matching full NFO" for a combination files, even if the file has URL (Universal Movie Scraper).

Code:

DEBUG: VideoInfoScanner: Found matching full NFO file: /data/media/Video/Movies/Besser geht’s nicht (1997)/Besser geht’s nicht (1997).DVD.nfo

Here my file "Besser geht’s nicht (1997).DVD.nfo":

xml:
<movie>

  <title>Besser geht’s nicht</title>

  <sorttitle>Besser geht's nicht</sorttitle>

</movie>

https://www.themoviedb.org/movie/2898-as...as-it-gets

This is just one example.
The file is UTF-8 (but I also tried ASCII/ANSI).
I also tried other URLs, e.g. just "http" (without 's'), shorter URL (ending with "movie/2898") and other things. I could not get rid of the problem.

The only solution for me – by now – is: Use the data as they are (if the scraper gets something) or create a full NFO file.

BTW: I have movies in folder, but only disc stubs (no ripped files, just list my DVDs and Blu-rays).

It's the latest official Kodi 18 on Linux, with the latest scraper from the official repo (Kodi 18.8, Universal Movie Scraper 5.4.0).


P.S.: Sorry for using this old thread, but it was the best matching my problem.

[Karellen]

I have split your post from the 2013 thread.

Unfortunately you cannot decide which scraper can be used on movie level (only on library level).

Yes you can. In exactly the same way you set a scraper for a Source, highlight the movie folder and set the scraper on a movie.

I am not really clear on what the problem is. You posted an example nfo file, what is wrong with it? Does it not work?

The scraped language is set by the scraper. Adding a link to a translation won't work as all the link does is point to the movie, it is the scraper that selects the language based on what your language setting is.

[User 421137]

I have split your post from the 2013 thread.

Well, okay. I'm also not very happy to warm up old threads, but it was the best I found (by Google, since it's hard or not possible to search in the forum with short term words like "URL" and "NFO").
(Beside it was not closed with a real solution or response if it worked then and on.)
I think it might be helpful to link also to the original thread: https://forum.kodi.tv/showthread.php?tid=162571
However, now it's in the correct (or at least in a better) sub forum.
 

Unfortunately you cannot decide which scraper can be used on movie level (only on library level).

Yes you can. In exactly the same way you set a scraper for a Source, highlight the movie folder and set the scraper on a movie.

Hm. Maybe I'm wrong!
If I go to "Movies" from the main menu I typically see my movies by "Title". If I go back ("..") I can enter it by "Recently added", "Genres", etc.
If I am in "Movies / Titles" I can select a movie and use e.g. the 'I' key to get details. In this info/details view I can make a "Refresh", but this is using the scraper, set for this source on library level.
If I am in "Movies / Titles" and I select a move and use the context menu I can select "Play", "Information", "Mark as watch", etc. and "Manage...". In the "Manage" menu I can set/change the title or sort title, link it to TV show, manage the set or remove it.
Only if I go into it by "Files" (by using ".." twice form "Movies / Titles", coming from the main menu option "Movies") / "Movies", then select the movie and use the context menu, I have the option "Change content". Then I get the same menu as on adding a video folder. I can also access it by "Setting" / "Media" / "Library" - "Videos..." / "Movies".
First: I didn't know this! For me it's also not very intuited to get this option only in this way. It's also a little bit confusing since it's not visible in the "Set content" windows for which level it appears.
It's also not very transparent to me if the "movie folder" is shown me as the movie title (e.g. "Besser geht’ nicht") and not as the real folder name (e.g. "Besser geht’s nicht (1997)").
For me this is not very transparent and intuitive at all.
However, if this works and I can set the scraper individually for each move, it will help me a little bit.
Nevertheless I think it would be fine to have the possibility to select the scraper directly at refresh and override the scraper, set at adding the whole movie folder as source.
If I could set a scraper also at move level (i.e. for the specific movie folder) and I can set different settings for each scraper, on each level, it gets hard to know where which settings are in use.
 

I am not really clear on what the problem is. You posted an example nfo file, what is wrong with it? Does it not work?

It was not working as expected! With my NFO file I only get a more or less empty move, only title and sort title have been set (and also the local poster and fanart was used), but all other data (plot, year, actors, etc.) was empty!
This seems to be because my NFO file was detected as "full NFO file" (in the log) and the URL was ignored. For me this is wrong. My NFO file is a combining file, with URL, and so the scraper should use the URL, scrap the data from the website of the URL and then override some data (in this case the title and the sorttitle).
 

The scraped language is set by the scraper. Adding a link to a translation won't work as all the link does is point to the movie, it is the scraper that selects the language based on what your language setting is.

This is clear to me.


P.S.: Default Skin "Estuary" is used, interface language is English, but scraper set to retrieve German data.

[Karellen]

It's also not very transparent to me if the "movie folder" is shown me as the movie title (e.g. "Besser geht’ nicht") and not as the real folder name (e.g. "Besser geht’s nicht (1997)").

Disable setting 2.2 here... https://kodi.wiki/view/Settings/Media/Videos#Files

Only if I go into it by "Files" (by using ".." twice form "Movies / Titles", coming from the main menu option "Movies") / "Movies", then select the movie and use the context menu, I have the option "Change content". Then I get the same menu as on adding a video folder. I can also access it by "Setting" / "Media" / "Library" - "Videos..." / "Movies".

Yes, this is the only section where you can set and adjust sources and scrapers. You cannot do this from Movies as this is Library, so there is no option to adjust files in Library. Use Videos>Files for this.

It was not working as expected! With my NFO file I only get a more or less empty move, only title and sort title have been set (and also the local poster and fanart was used), but all other data (plot, year, actors, etc.) was empty!

I tested your nfo above and it works fine for me here. This is in my log...

xml:
DEBUG: scraper: NfoUrl returned <details><url>https://api.tmdb.org/3/movie/2898?api_key=f090bb54758cabf231fb605d3e3e0468&amp;language=en</url><id>2898</id></details>

DEBUG: VideoInfoScanner: Found matching mixed NFO file: C:\KODI\Test- Movies\As Good as It Gets (1997)\As Good as It Gets (1997).nfo

DEBUG: VideoInfoScanner: Fetching url 'https://api.tmdb.org/3/movie/2898?api_key=f090bb54758cabf231fb605d3e3e0468&language=en' using The Movie Database scraper (content: 'movies')

DEBUG: Thread VideoInfoDownloader start, auto delete: false

If you provide a Debug Log which captures you scraping the movie again, I might see what the problem is.

[User 421137]

It's also not very transparent to me if the "movie folder" is shown me as the movie title (e.g. "Besser geht’ nicht") and not as the real folder name (e.g. "Besser geht’s nicht (1997)").

Disable setting 2.2 here... https://kodi.wiki/view/Settings/Media/Videos#Files

Oh! Thx!

I tested your nfo above and it works fine for me here. This is in my log...

xml:
DEBUG: scraper: NfoUrl returned <details><url>https://api.tmdb.org/3/movie/2898?api_key=f090bb54758cabf231fb605d3e3e0468&amp;language=en</url><id>2898</id></details>

DEBUG: VideoInfoScanner: Found matching mixed NFO file: C:\KODI\Test- Movies\As Good as It Gets (1997)\As Good as It Gets (1997).nfo

DEBUG: VideoInfoScanner: Fetching url 'https://api.tmdb.org/3/movie/2898?api_key=f090bb54758cabf231fb605d3e3e0468&language=en' using The Movie Database scraper (content: 'movies')

DEBUG: Thread VideoInfoDownloader start, auto delete: false

If you provide a Debug Log which captures you scraping the movie again, I might see what the problem is.

I'll try it later, when I'm back on my machine but I saw nothing interesting in my log, just the message "DEBUG: VideoInfoScanner: Found matching full NFO file ...". Maybe I have to activate some additional logging (component specific?).
Your test differs a little bit. You are on Windows, I use Linux. Your directory and filename are different: I use the German title, with correct apostrophe and also with a media flag.

[Karellen]

aybe I have to activate some additional logging (component specific?).

No

Your directory and filename are different: I use the German title, with correct apostrophe and also with a media flag.

Should not make a difference. It is the nfo file I was testing, not the name scrape. I could name it anything and the nfo should still be found and handled correctly.

[User 421137]

Now I tried it again. I started Kodi with activated Debug and made a refresh of the movie (with a combining file NFO file).
In the log file there is again just the strange "Found matching full NFO file":

Code:

$ cat ~/.kodi/temp/kodi.old.log | grep -n -C 1 Besser
2452-2020-09-03 20:59:50.247 T:139875074886336 DEBUG: CAddonSettings[metadata.universal]: loading setting values
2453:2020-09-03 20:59:50.258 T:139875074886336 DEBUG: VideoInfoScanner: Found matching full NFO file: /data/media/Video/Movies/Besser geht’s nicht (1997)/Besser geht’s nicht (1997).DVD.nfo
2454-2020-09-03 20:59:50.259 T:139875074886336 DEBUG: ------ Window Init (DialogConfirm.xml) ------
--
2468-2020-09-03 20:59:52.144 T:139875074886336 DEBUG: CAddonSettings[metadata.universal]: loading setting values
2469:2020-09-03 20:59:52.172 T:139875074886336 DEBUG: VideoInfoScanner: Found matching full NFO file: /data/media/Video/Movies/Besser geht’s nicht (1997)/Besser geht’s nicht (1997).DVD.nfo
2470:2020-09-03 20:59:52.173 T:139875074886336 DEBUG: VideoInfoScanner: Adding new item to movies:/data/media/Video/Movies/Besser geht’s nicht (1997)/Besser geht’s nicht (1997).DVD.disc
2471-2020-09-03 20:59:52.187 T:139875074886336 DEBUG: ------ Window Init (DialogVideoInfo.xml) ------

My NFO file (and other movie files):

Code:

$ ll "/data/media/Video/Movies/Besser geht’s nicht (1997)/"
total 332
drwxr-xr-x   2 nero nero   4096 Sep  3 20:58  ./
drwxr-xr-x 127 nero nero  12288 Sep  1 20:55  ../
-rw-r--r--   1 nero nero     71 Aug 29 07:21 'Besser geht’s nicht (1997).DVD.disc'
-rw-r--r--   1 nero nero    158 Aug 29 07:21 'Besser geht’s nicht (1997).DVD.nfo'
-rw-r--r--   1 nero nero 247411 Sep  1 04:29  fanart.jpg
-rw-r--r--   1 nero nero  62095 Aug 29 04:07  poster.jpg
$
$
$ cat "/data/media/Video/Movies/Besser geht’s nicht (1997)/Besser geht’s nicht (1997).DVD.nfo"
<movie>
  <title>Besser geht’s nicht</title>
  <sorttitle>Besser geht's nicht</sorttitle>
</movie>
https://www.themoviedb.org/movie/2898-as...as-it-gets
$

Full log: paste.kodi.tv


P.S.: I don't know why the URL is displayed shortened in the syntax block. It's of course the full valid URL https://www.themoviedb.org/movie/2898-as...as-it-gets in the NFO file.

[User 421137]

Maybe also the scraper settings is helpful: ~/.kodi/userdata/addon_data/metadata.universal/settings.xml

[Karellen]

Ok, the Debug Log revealed the problem.

You have this link in your nfo file... https://www.themoviedb.org/movie/2898-as...as-it-gets which is a link from TheMovieDB

But you are using the Universal Movie Scraper which uses IMDB. Change your link to the IMDB URL.

[User 421137]

Ok, the Debug Log revealed the problem.

You have this link in your nfo file... https://www.themoviedb.org/movie/2898-as...as-it-gets which is a link from TheMovieDB

But you are using the Universal Movie Scraper which uses IMDB. Change your link to the IMDB URL.

Hm, I think this should not be the problem because the Universal Movie Scraper also supports TMDB.

Add-on:Universal_Movie_Scraper: "... supported sites: IMDb, themoviedb.org, "

[Karellen]

Ok. You must be the expert, so I have no other suggestions.

[User 421137]

I also checked the scraper itself and it supports TMDB.

Code:

$ grep -m 1 themoviedb ~/.kodi/addons/metadata.universal/universal.xml
<expression clear="yes" noclean="1">(themoviedb.org/movie/)([0-9]*)</expression>

It's quite hard to read all the RegEx but it basically looks good and it should support my combining NFO with a TMDB link.

For several reasons I'd like to have TMDB (for all of my movies).

You, Karellen, also tested it with my NFO file, having a TMDB link and it was OK!?

[Karellen]

You, Karellen, also tested it with my NFO file, having a TMDB link and it was OK!?

Yes, I tested with TheMovieDB scraper. Of course you never told me you were using the Universal Movie Scraper otherwise I would have spotted your mistake immediately.

If you want to persist with TheMovieDB URL's, then go the the Universal Movie Scraper settings, and under the Search tab, change the Search Engine To Use to TheMovieDB. Make this change at your Source by selecting Change content from the Context menu, then selecting Settings under the Universal Movie Scraper entry.

[User 421137]

you never told me you were using the Universal Movie Scraper

It's in the initial post :

I get "Found matching full NFO" for a combination files, even if the file has URL (Universal Movie Scraper).
...
It's the latest official Kodi 18 on Linux, with the latest scraper from the official repo (Kodi 18.8, Universal Movie Scraper 5.4.0).

 

If you want to persist with TheMovieDB URL's, then go the the Universal Movie Scraper settings, and under the Search tab, change the Search Engine To Use to TheMovieDB. Make this change at your Source by selecting Change content from the Context menu, then selecting Settings under the Universal Movie Scraper entry.

Well, yes, it looks this was the problem.

With

Code:

$ grep searchservice ~/.kodi/userdata/addon_data/metadata.universal/settings.xml
<setting id="searchservice">themoviedb.org</setting>

it is working.
(This is just to be the global settings for the scraper but it is like in the DB [path.strSetting])

I thought I had set everything to TMDB but I did it only till the tab "Rating" (global settings and also on source level). Maybe this was because I didn't see the "Search" tab, since I have to scroll down (but also for "Rating" I had to scroll ...).
Maybe it might be fine to have such an important setting on top.
Maybe also the UI would be helpful, if it indicates that there are more options and you have to scroll down.
I also wonder somehow why the setting for the search engine is necessary for a combination NFO file, since the included URL could/should force the used search engine (even if this setting might be needed in other cases). Maybe this is a limitation of the scraper framework but it would be fine not to be limited to global (or source/folder) settings here.

At all, I think the handling may be improved.
Maybe a XML declaration should be really mandatory for a full (non-combination) NFO file. Such a combination NFO file should also not be logged as "full NFO file", since it's clearly not a full NFO (e.g. missing required fields) and the URL makes it to a combination file. This wrong logging info causes all the confusion ...

However, it looks good now (as I can see on a first test).

Thx!